Sound matching algorithms seek to approximate a target waveform by parametric audio synthesis. Deep neural networks have achieved promising results in matching sustained harmonic tones. However, the task is more challenging when targets are nonstationary and inharmonic, e.g., percussion. We attribute this problem to the inadequacy of loss function. On one hand, mean square error in the parametric domain, known as "P-loss", is simple and fast but fails to accommodate the differing perceptual significance of each parameter. On the other hand, mean square error in the spectrotemporal domain, known as "spectral loss", is perceptually motivated and serves in differentiable digital signal processing (DDSP). Yet, spectral loss has more local minima than P-loss and its gradient may be computationally expensive; hence a slow convergence. Against this conundrum, we present Perceptual-Neural-Physical loss (PNP). PNP is the optimal quadratic approximation of spectral loss while being as fast as P-loss during training. We instantiate PNP with physical modeling synthesis as decoder and joint time-frequency scattering transform (JTFS) as spectral representation. We demonstrate its potential on matching synthetic drum sounds in comparison with other loss functions.
Climate change is expected to intensify and increase extreme events in the weather cycle. Since this has a significant impact on various sectors of our life, recent works are concerned with identifying and predicting such extreme events from Earth observations. This paper proposes a 2D/3D two-branch convolutional neural network (CNN) for wildfire danger forecasting. To use a unified framework, previous approaches duplicate static variables along the time dimension and neglect the intrinsic differences between static and dynamic variables. Furthermore, most existing multi-branch architectures lose the interconnections between the branches during the feature learning stage. To address these issues, we propose a two-branch architecture with a Location-aware Adaptive Denormalization layer (LOADE). Using LOADE as a building block, we can modulate the dynamic features conditional on their geographical location. Thus, our approach considers feature properties as a unified yet compound 2D/3D model. Besides, we propose using an absolute temporal encoding for time-related forecasting problems. Our experimental results show a better performance of our approach than other baselines on the challenging FireCube dataset.
Modern Deep Reinforcement Learning (RL) algorithms require estimates of the maximal Q-value, which are difficult to compute in continuous domains with an infinite number of possible actions. In this work, we introduce a new update rule for online and offline RL which directly models the maximal value using Extreme Value Theory (EVT), drawing inspiration from Economics. By doing so, we avoid computing Q-values using out-of-distribution actions which is often a substantial source of error. Our key insight is to introduce an objective that directly estimates the optimal soft-value functions (LogSumExp) in the maximum entropy RL setting without needing to sample from a policy. Using EVT, we derive our Extreme Q-Learning framework and consequently online and, for the first time, offline MaxEnt Q-learning algorithms, that do not explicitly require access to a policy or its entropy. Our method obtains consistently strong performance in the D4RL benchmark, outperforming prior works by 10+ points on some tasks while offering moderate improvements over SAC and TD3 on online DM Control tasks.
Accurate whole-body multi-person pose estimation and tracking is an important yet challenging topic in computer vision. To capture the subtle actions of humans for complex behavior analysis, whole-body pose estimation including the face, body, hand and foot is essential over conventional body-only pose estimation. In this paper, we present AlphaPose, a system that can perform accurate whole-body pose estimation and tracking jointly while running in realtime. To this end, we propose several new techniques: Symmetric Integral Keypoint Regression (SIKR) for fast and fine localization, Parametric Pose Non-Maximum-Suppression (P-NMS) for eliminating redundant human detections and Pose Aware Identity Embedding for jointly pose estimation and tracking. During training, we resort to Part-Guided Proposal Generator (PGPG) and multi-domain knowledge distillation to further improve the accuracy. Our method is able to localize whole-body keypoints accurately and tracks humans simultaneously given inaccurate bounding boxes and redundant detections. We show a significant improvement over current state-of-the-art methods in both speed and accuracy on COCO-wholebody, COCO, PoseTrack, and our proposed Halpe-FullBody pose estimation dataset. Our model, source codes and dataset are made publicly available at https://github.com/MVIG-SJTU/AlphaPose.
Artificial Intelligence (AI) is transforming all fields of knowledge and production. From surgery, autonomous driving, to image and video creation, AI seems to make possible hitherto unimaginable processes of automation and efficient creation. Media and communication are not an exception, and we are currently witnessing the dawn of powerful AI tools capable of creating artistic images from simple keywords, or to capture emotions from facial expression. These examples may be only the beginning of what can be in the future the engines for automatic AI real time creation of media content linked to the emotional and behavioural responses of individuals. Although it may seem we are still far from there, it is already the moment to adapt our theories about media to the hypothetical scenario in which content production can be done without human intervention, and governed by the controlled any reactions of the individual to the exposure to media content. Following that, I propose the definition of the Integrated Model of Artificial Intelligence-Mediated Communication Effects (IMAGINE), and its consequences on the way we understand media evolution (Scolari, 2012) and we think about media effects (Potter, 2010). The conceptual framework proposed is aimed to help scholars theorizing and doing research in a scenario of continuous real-time connection between AI measurement of people's responses to media, and the AI creation of content, with the objective of optimizing and maximizing the processes of influence. Parasocial interaction and real-time beautification are used as examples to model the functioning of the IMAGINE process.
Distributed deep learning (DDL) systems strongly depend on network performance. Current electronic packet switched (EPS) network architectures and technologies suffer from variable diameter topologies, low-bisection bandwidth and over-subscription affecting completion time of communication and collective operations. We introduce a near-exascale, full-bisection bandwidth, all-to-all, single-hop, all-optical network architecture with nanosecond reconfiguration called RAMP, which supports large-scale distributed and parallel computing systems (12.8~Tbps per node for up to 65,536 nodes). For the first time, a custom RAMP-x MPI strategy and a network transcoder is proposed to run MPI collective operations across the optical circuit switched (OCS) network in a schedule-less and contention-less manner. RAMP achieves 7.6-171$\times$ speed-up in completion time across all MPI operations compared to realistic EPS and OCS counterparts. It can also deliver a 1.3-16$\times$ and 7.8-58$\times$ reduction in Megatron and DLRM training time respectively} while offering 42-53$\times$ and 3.3-12.4$\times$ improvement in energy consumption and cost respectively.
Recent studies have shown great promise in applying graph neural networks for multivariate time series forecasting, where the interactions of time series are described as a graph structure and the variables are represented as the graph nodes. Along this line, existing methods usually assume that the graph structure (or the adjacency matrix), which determines the aggregation manner of graph neural network, is fixed either by definition or self-learning. However, the interactions of variables can be dynamic and evolutionary in real-world scenarios. Furthermore, the interactions of time series are quite different if they are observed at different time scales. To equip the graph neural network with a flexible and practical graph structure, in this paper, we investigate how to model the evolutionary and multi-scale interactions of time series. In particular, we first provide a hierarchical graph structure cooperated with the dilated convolution to capture the scale-specific correlations among time series. Then, a series of adjacency matrices are constructed under a recurrent manner to represent the evolving correlations at each layer. Moreover, a unified neural network is provided to integrate the components above to get the final prediction. In this way, we can capture the pair-wise correlations and temporal dependency simultaneously. Finally, experiments on both single-step and multi-step forecasting tasks demonstrate the superiority of our method over the state-of-the-art approaches.
We explore how observational and interventional causal discovery methods can be combined. A state-of-the-art observational causal discovery algorithm for time series capable of handling latent confounders and contemporaneous effects, called LPCMCI, is extended to profit from casual constraints found through randomized control trials. Numerical results show that, given perfect interventional constraints, the reconstructed structural causal models (SCMs) of the extended LPCMCI allow 84.6% of the time for the optimal prediction of the target variable. The implementation of interventional and observational causal discovery is modular, allowing causal constraints from other sources. The second part of this thesis investigates the question of regret minimizing control by simultaneously learning a causal model and planning actions through the causal model. The idea is that an agent to optimize a measured variable first learns the system's mechanics through observational causal discovery. The agent then intervenes on the most promising variable with randomized values allowing for the exploitation and generation of new interventional data. The agent then uses the interventional data to enhance the causal model further, allowing improved actions the next time. The extended LPCMCI can be favorable compared to the original LPCMCI algorithm. The numerical results show that detecting and using interventional constraints leads to reconstructed SCMs that allow 60.9% of the time for the optimal prediction of the target variable in contrast to the baseline of 53.6% when using the original LPCMCI algorithm. Furthermore, the induced average regret decreases from 1.2 when using the original LPCMCI algorithm to 1.0 when using the extended LPCMCI algorithm with interventional discovery.
Compared with CINE phase contrast MRI (CINE-PC), echo-planar imaging phase contrast (EPI-PC) can achieve realtime quantification of blood flow, with lower SNR. In this study, the pulsating real model of the simulated cerebral vasculature was used to verify the accuracy of EPI-PC. The imaging time of EPI-PC was 62ms/image at 100*60 spatial resolution. The reconstructed EPI-PC flow curve was extracted by homemade post-processing software. After comparison with the CINE-PC flow curve, it was concluded that EPI-PC can provide an average flow with less than 3% error, and its flow curve will be similar to the CINE-PC flow curve in shape.
In this paper, we propose an AI based approach to Trailer Generation in the form of short videos for online educational courses. Trailers give an overview of the course to the learners and help them make an informed choice about the courses they want to learn. It also helps to generate curiosity and interest among the learners and encourages them to pursue a course. While it is possible to manually generate the trailers, it requires extensive human efforts and skills over a broad spectrum of design, span selection, video editing, domain knowledge, etc., thus making it time-consuming and expensive, especially in an academic setting. The framework we propose in this work is a template based method for video trailer generation, where most of the textual content of the trailer is auto-generated and the trailer video is automatically generated, by leveraging Machine Learning and Natural Language Processing techniques. The proposed trailer is in the form of a timeline consisting of various fragments created by selecting, para-phrasing or generating content using various proposed techniques. The fragments are further enhanced by adding voice-over text, subtitles, animations, etc., to create a holistic experience. Finally, we perform user evaluation with 63 human evaluators for evaluating the trailers generated by our system and the results obtained were encouraging.